SPE 76729

A Simulation Study of Novel Thermal Recovery Methods in the Ugnu Tar Sand
Reservoir, North Slope, Alaska

Bakul C. Sharma, Santanu Khataniar, SPE, Shirish L. Patil, SPE, Vidyadhar A. Kamath, SPE, and Abhijit Y. Dandekar,
SPE, University of Alaska Fairbanks

Copyright 2002, Society of Petroleum Engineers Inc.

place. The Ugnu sands are highly heterogeneous with large
This paper was prepared for presentation at the SPE Western Regional/AAPG Pacific Section variations in porosity, permeability, water saturation and oil
Joint Meeting held in Anchorage, Alaska, U.S.A., 20–22 May 2002.
viscosity. The Ugnu crude is highly viscous bitumen with
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
viscosity ranging from 50,000 to 10 million centipoise at
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to reservoir conditions. Because of such high viscosity,
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at reasonable oil production rates cannot be achieved without
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
some form of thermal enhanced oil recovery (EOR) method,
for commercial purposes without the written consent of the Society of Petroleum Engineers is usually steam injection. A common problem associated with
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous steam injection in tar sand and bitumen is steam injectivity.
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Even with horizontal wells, initial steam cycling along with
fracturing may be necessary to establish steam injectivity.
Due to advances in the technology of drilling horizontal
Abstract
wells, the cost to benefit ratio of horizontal wells is
The Ugnu reservoir, located in the Kuparuk River Unit on the
decreasing. As a result, horizontal wells are being
North Slope of Alaska, is estimated to contain over 6 billion
increasingly used in EOR projects, specially in thermal oil
barrels of heavy oil in the form of tar and bitumen. Due to the
recovery. Horizontal wells have been drilled in pilot projects
extreme oil viscosities conventional enhanced oil recovery
for thermal recovery in bitumen and heavy oil reservoirs. The
(EOR) methods are not readily applicable to the Ugnu sands.
main advantages of the use of horizontal wells over vertical
Additionally, high reservoir heterogeneity creates
wells are improved sweep efficiency, increased recoverable
complications with EOR processes. A detailed simulation
reserves, increased steam injectivity, and fewer number of
study was conducted to evaluate the performance of four novel
wells required for field development1. A 2000 to 4000-ft long
thermal recovery methods in the Ugnu reservoir. These
horizontal well can replace several vertical well, reducing
methods included steam assisted gravity drainage (SAGD)
investment and operational costs. With the initial cost of
with horizontal injectors, SAGD with vertical injectors, SAGD
drilling a horizontal well on the decline, horizontal wells
with electrical preheating (EP-SAGD), and EP-SAGD with
appear to be an excellent choice for EOR operations.
vertical injectors. A conventional steamflood was also
Thermal recovery processes in heavy oil or tar deposits
simulated for comparison as the base case. The effects of
with negligible initial oil mobility require steam injection to
various reservoir and process parameters such as well spacing,
establish interwell communication to unlock the reservoir.
flood pattern, reservoir heterogeneity, and anisotropy on
The existence of very large heavy oil and tar sand reservoirs
process performance were studied. The four SAGD based
without significant steam injectivity has led to the
processes were found to be far superior to conventional
development of a novel thermal recovery method known as
steamflood, which performed very poorly. An optimum
the steam assisted gravity drainage (SAGD) method2. In the
spacing between injector and producer for SAGD methods
typical SAGD process, horizontal steam injection wells and oil
was determined. The EP-SAGD process was found to be
producing wells are placed close to each other, with the
thermally more efficient than SAGD, however, final oil
injection well located above the production well. The
recoveries were comparable for the two methods. Placement
injection well is placed near the bottom of the reservoir. A
of horizontal producer was found to be more critical than
minimum pressure differential, close to the gravity head
placement of the injector.
differential, is desirable between the injection and production
Introduction wells. The injected steam travels from the bottom of the
The Ugnu tar sands located in the Kuparuk River Unit on the reservoir upwards creating a rising steam chamber. The hot
North Slope of Alaska contain over six billion barrels of oil in displaced oil and condensate are drained by gravity and
2 B. SHARMA, S. KHATANIAR, S. PATIL, V. KAMATH AND A. DANDEKAR
produced through the bottom horizontal (production) well.
The main advantage of the process over a conventional steam
drive is that hot oil is produced by gravity drainage as soon as
it is displaced from the reservoir. As steam condenses at the
interface of the steam chamber and cold reservoir, heat is
transferred to the reservoir, the oil near the interface is heated
and drained to the production well. Thus, the steam chamber
expands vertically upwards initially and then sideways,
significantly increasing sweep efficiency. It is also possible to
use vertical injection wells instead of horizontal wells. For a
successful gravity drainage process, the reservoir should be
fairly thick (50 ft or more). Good vertical permeability is also
necessary. Continuous shale barriers could impede the
process efficiency, but small, discontinuous shale barriers may
not have a significant effect. Experiments by Chung and
Butler3 with two-dimensional scaled models have shown the
ultimate oil recovery to be independent of well spacing,
however, close well spacing provided higher rates of
oil recovery.
One of the most serious obstacles in thermal oil
recovery of tar sand and bitumen is the lack of steam
injectivity. Electrical heating offers a technically viable
option, which has been the subject of studies and field tests.
Due to its high cost, electrical heating is not viewed as a
replacement for steam injection, but rather a precursor to
steam injection. Thus, the SAGD process can be preceded by
a period of electrical heating to improve steam injectivity.
Such a process will be referred to as the electrically preheated
SAGD or EP-SAGD method. In the EP-SAGD method, three
parallel horizontal wells are placed triangularly in a vertical
cross section. The top well is the injector and the two bottom
wells serve as the producers. In addition, the wells also serve
as the electrodes to preheat the formation electrically. At the
end of the preheating period, steam injection and fluid
production are initiated. The preheated area inside and around
the well's triangle reaches low tar viscosities and is readily
produced. This results in greater recovery than the standard
SAGD process. Another advantage of the EP-SAGD method
is its relative insensitivity to the presence of shale barriers, as
shown by the experimental work of Glandt and Hsu4.
Because oil production from the State of Alaska has
been on the decline, there is need for development of EOR
techniques for oil production form the heavy oil and tar sand
resources in the state. The main objective of this research was
to evaluate some novel thermal recovery methods which may
be applicable to the recovery of the highly viscous Ugnu
crude. A thermal simulator was used to perform three-
dimensional simulations of conventional steamflood, SAGD,
SAGD with vertical injectors, EP-SAGD, and EP-SAGD with
vertical injectors in the Ugnu reservoir. The effect of various
reservoir and process parameters on the performance of these
methods was studied.
SPE 76729
The Ugnu Sands
Detailed description of reservoir and geological characteristics
of the Ugnu tar sands is available in the literature5,6. The ugnu
sands overlie the West Sak and the Kuparuk sands on the
Alaskan North Slope, with the potentially steam developable
section located on the northern part of the Kuparuk River Unit.
The Ugnu sands are shallow with thick pay intervals occurring
at depths ranging from 2200 ft to 3200 ft. Because of the
shallow depth, reservoir temperature and hence, oil viscosity
are influenced by the overlying layer of permafrost. The depth
of permafrost varies from 1200 to 1700 ft. The Ugnu sands
exhibit good porosity ranging from 34% to 37%. Measured
air permeabilities range from several hunderd millidarcies to
several darcies. Oil saturation estimated from log and core
analysis ranges from 66% to 72%. Reservoir temperature is
fairly low, 45oF in the west to 65oF in the east. Reservoir
pressure is also low (1328 psi at 2978 ft), which makes these
sands good candidate for steam injection. Viscosity of dead
oil at reservoir temperature varies from 100,000 cp in the east
to 15 million cp in the west. Oil API gravity varies from 7
to 11.5.
Method of Study
A detailed simulation study was conducted to evaluate and
compare the performance of the following thermal recovery
methods in the Ugnu reservoir.
(a) Conventional steamflood
(b) Steam assisted gravity drainage (SAGD)
(c) Electrically preheated SAGD (EP-SAGD)
(d) SAGD with vertical injectors
(e) EP-SAGD with vertical injectors
Because reservoir properties vary significantly within the
Ugnu sands, a sensitivity analysis was performed by varying
the following reservoir and process parameters.
(a) Vertical well spacing
(b) Electric preheating time
(c) Vertical injectors
(d) Lateral well spacing
(e) Reservoir anisotropy
(f) Reservoir heterogeneity
The tasks listed above were performed by using a commercial
three-phase, multi-component thermal and steam additive
simulator. The simulations were run in the adaptive implicit
mode with the exception of the well gridblocks, which were
solved fully implicitly.
Reservoir input data used in these simulations are
shown in Table 1. Table 1 includes the base case values as
well as the parameter ranges over which sensitivity analysis
was performed. These data are based on previous work by
Sinha7, Zhang8 and Hallam et al.5. Relative permeability data
were obtained from the field operator. A correlation for
variation of the Ugnu oil viscosity with temperature,
developed by Sinha7, was used in this study. The correlation
incorporated the effect of dissolved gas on oil viscosity.
 A SIMULATION STUDY OF NOVEL THERMAL RECOVERY
SPE 76729 METHODS IN THE UGNU TAR SAND RESERVOIR, NORTH SLOPE, ALASKA
Results and Discussion
Base Case Simulations. A homogeneous system with the
average values of reservoir properties within the potentially
developable section of the Ugnu reservoir was used for
simulating the base case for each of the thermal processes
studied. The base case used a model reservoir pattern of 90 ft
thickness, 1000 ft length and 200 ft width, and the simulation
was conducted for 5000 days of production. For the
conventional SAGD and EP-SAGD methods, a horizontal
producer of 1000 ft length was placed 5 ft above the bottom of
the reservoir. A horizontal injector of same length was placed
parallel to the producer 10 ft above it (Figure 1). Other
parameters used in the base case are shown in Table 1. In the
EP-SAGD method, electrical preheating was done for
365 days.
A grid refinement study was first performed for the
SAGD and EP-SAGD methods. The following grid patterns
were considered: (i) 9 blocks (3x1x3), (ii) 18 blocks (3x1x6),
(iii) 90 blocks (3x5x6), and (iv) 180 blocks (6x5x6). The final
oil recovery after 5000 days was the same for all four grids.
However, oil recovery as a function of time and oil production
rate as a function of time were significantly different for the 9,
18 and 90 gridblock systems. The 90 and 180 gridblock
systems gave similar results. Therefore, the 180 gridblock
system was selected for performing all the simulations.
A conventional steamflood was simulated using two
vertical injectors and two vertical producers. This resulted in
an oil recovery of only 136 STB (less than 0.1% of IOIP) after
5000 days of steam injection. Because of such poor
performance, conventional steamflood was not studied further.
The base case for SAGD and EP-SAGD with vertical
injectors was studied using 1, 2, and 4 vertical injection wells
as shown in Figure 2. Oil recovery from the 1-injector case
was much lower than the 2-injector case. With 4 injectors, oil
was produced faster initially, however, cumulative recovery
after 5000 days was the same as that for the 2-injector case.
Since using 4 injectors would increase initial investment
significantly, further studies were continued with the 2 vertical
injectors. A comparison of cumulative oil recovery as a
function of time for the base case of SAGD, EP-SAGD,
SAGD with two vertical injectors and EP-SAGD with two
vertical injectors is shown in Figure 3. The EP-SAGD process
shows a lag in oil recovery compared to the SAGD process
due to the preheating period, however, the final recovery after
5000 days is nearly 70% of IOIP for all of these methods.
Effect of Process Parameters. The effects of various process
parameters on oil recovery are discussed next.
Injector-Producer Vertical Spacing. In the base case,
the horizontal producer was placed in the bottom layer, i.e., 5
ft from the bottom of the reservoir and the injector was placed
10 ft above the producer. To study the effect of injector to
producer spacing on oil recovery, the position of the injector
was varied from 10 to 70 ft above the producer. The oil
recovery plots for SAGD and EP-SAGD with varying well
3
spacing are shown in Figures 4 and 5, respectively.
Cumulative oil recovery after 5000 days remained unaffected
by the injector to producer spacing, however, the 30 and 45 ft
spacing resulted in higher initial recovery than the base case as
well as the 70 ft spacing. In addition, the 70 ft spacing
resulted in negligible oil recovery in the first 500 to 700 days.
This indicates that for a given reservoir, there exists an
optimum vertical spacing between the injector and the
producer for achieving the fastest oil recovery.
Duration of Electrical Preheating. Simulation runs
were carried out for 200, 365, 730,1095 and 2000 days of
preheating time for the EP-SAGD process. The injector
temperature was maintained at 574oF during preheating and no
steam was injected. Figure 6 shows that there is an increasing
lag in oil production as the preheating time increases.
However, once steam injection begins, oil recovery rates are
about the same for all preheating times. Cumulative oil
recovery at 5000 days is slightly less for the case of 2000 days
of preheating due to the initial lag in production.
Lateral Well Spacing. The lateral well spacing was
varied from 100 ft to 400 ft in steps of 100 ft for the SAGD
process. Figure 7 is cumulative oil recovery as a function of
time when steam injection rate per injector was fixed at 1000
barrels/day. Figure 7 shows that the rate of oil recovery
increases as the well spacing is decreased. Therefore, a high
recovery rate can be achieved by using a series of long
horizontal production wells parallel to each other. The final
recovery after 5000 days is the same for 100 ft and 200 ft
spacing, but decreases significantly for the higher spacings.
When the total steam injection to the system is kept constant,
it is observed that a spacing of 200 ft gives the highest oil
recovery rate (Figure 8). Thus, there exists an optimum lateral
spacing for the fastest oil recovery.
Anisotropy. In order to study the effect of anisotropy,
the ratio of vertical to horizontal permeability was varied from
0.15 to 1. Figure 9 shows the effect of anisotropy on oil
production rate for SAGD process. The peak production rate
decreased significantly as the permeability ratio dropped from
1 to 0.15. This is because the steam rising rate and the oil
drainage rate both are impaired by vertical permeability
reduction. Figure 10 shows the final cumulative oil recovery
at 5000 days as a function of permeability ratio for SAGD,
EP-SAGD and SAGD with vertical injectors. For
permeability ratio of 1, all three processes resulted in nearly
the same final recovery. When the permeability ratio dropped
below 0.6, there was marked decrease in the recovery for all
three processes. The SAGD with vertical injectors was the
most affected process.
Heterogeneity. To study the effect of heterogeneity,
the reservoir was divided into two layers. The top layer was
50 ft thick and the bottom layer was 40 ft thick. Three
simulation runs were performed for each of the recovery
methods: SAGD, EP-SAGD and SAGD with vertical
injectors. In the first run, both layers were assigned the same
permeability value of 680 md. In the second run, the bottom
layer was assigned a permeability of 680 md and the top layer
4 B. SHARMA, S. KHATANIAR, S. PATIL, V. KAMATH AND A. DANDEKAR SPE 76729

was assigned a permeability of 100 md. In the third run, 680 References
md was the permeability of the top layer and 100 md was 1. Joshi, S. D.: “Thermal Oil Recovery with Horizontal Wells,” J.
assigned to the bottom layer. Final oil recoveries after 5000 Pet. Tech., Nov. 1991, Vol. 43, No. 11, p. 1302-1304.
days for the three methods and the three permeability ratios 2. Butler, R. M. and Stephens, D. J.: “The Gravity Drainage of
are shown in Figure 11. Reduced permeability in the top layer Steam-Heated Oil to Parallel Horizontal Wells,” J. Can. Pet.
Tech., April-June 1981, p. 90-96.
(second run) resulted in only a slight reduction in oil recovery 3. Chung, K. H. and Butler, R. M.: “A Theoretical and
compared to the homogeneous case. SAGD with vertical Experimental Study of SAGD Process,” paper #89, Proc. 4th
injectors performed poorly due to reduced injectivity in the top UNITAR/UNDP International Conference on Heavy Crude and
layer. Reduced permeability in the bottom layer (third run) Tar Sands, Edmonton, Canada, Aug. 7-12, 1988, p. 191-210.
resulted in drastic reduction of oil recovery by all three 4. Glandt, C. A. and Hsu, C.: “Electrical Preheating in Low
methods due to poor drainage by the producing well located in Injectivity Tar Sand Deposits,” Proc. 8th SPE/DOE Symposium
the low permeability layer. Cumulative oil recovery as a on Enhanced Oil Recovery, Tulsa, OK, April 22-24, 1992,
function of time for the SAGD method is shown in Figure 12. p. 91-98.
It is clear from this figure that continuing steam injection 5. Hallam, R. J., Abou-Sayed, A. S., Garon, A. M., Putman, T. W.
and Weggeland, M. C.: “Resource Description and
beyond 5000 days will not significantly increase the final oil Development Potential of the Ugnu Reservoir, North Slope,
recovery. However, when vertical injectors are used with the Alaska,” paper SPE 21779, Proc. SPE Western Regional
SAGD method, oil recovery continues to build at 5000 days in Meeting, Long Beach, CA, Mar. 20-22, 1991, p. 285-300.
the heterogeneous cases, so that further increase in recovery 6. Sharma, G. D., Ogbe, D. O., Kamath, V. A. and Zhang, M.:
can be expected (Figure 13). Thus, reservoir heterogeneity is “Shallow Sands of North Slope, Alaska and their Hydrocarbon
very important in determining placement of the wells and Potentials,” in Particle Technology and Surface Phenomena in
choice of injector type (vertical vs. horizontal). Minerals and Petroleum, Plenum Publishing Co., NY, 1991,
p. 105-134.
Conclusions 7. Sinha, S.: “Numerical Simulation Study of Steam Assisted
1. For the base case with average rock and fluid properties of Gravity Drainage Process in Ugnu Tar Sand Reservoir,” MS
the Ugnu sands, all SAGD and EP-SAGD methods (horizontal Thesis, University of Alaska Fairbanks, Aug. 1992.
8. Zhang, M.: “Characterization and Description of Lower Ugnu
and vertical injectors) gave about the same final oil recovery and West Sak Reservoirs using Well Logs,” MS Thesis,
of nearly 70% after 5000 days of injection. Thus, SAGD University of Alaska Fairbanks, Dec. 1989.
appears to be a suitable method for recovering the Ugnu crude.
Conventional steamflood produced unacceptable results. SI Metric Conversion Factors
2. There exists an optimum vertical spacing between injector bbl x 1.589 873 E-01 = m3
and producer for maximizing recovery by SAGD and EP- cp x 1.0* E-03 = Pa.s
SAGD methods. darcy x 9.869 233 E-13 = m2
3. Oil recoveries with vertical injectors for base case ft x 3.048* E-01 = m
psi x 6.894 757 E+00 = kPa
simulation are comparable to those obtained using horizontal
*Conversion factor is exact
injectors for both SAGD and EP-SAGD processes. This is
because of the assumption of isotropic formation in the Table 1
base case. Ranges of Reservoir Properties within Ugnu Sands and Data Used for Base
4. Longer period of electrical preheating increases oil steam Case Ugnu Reservoir Simulation
ratio, but delays oil recovery. Final recoveries for the SAGD Properties Range Values
and EP-SAGD processes were nearly the same. Electrical Porosity 0.27-0.39 0.37
preheating marginally increases recovery per unit heat input. Permeability (md) 35-1500 680
Water Saturation 0.28-0.34 0.28
Thus, EP-SAGD is thermally somewhat more efficient than Reservoir Gross Thickness (ft) 75-105 90
SAGD process. Reservoir Temperature (°F) 45-70 65
5. The final oil recovery is independent of the lateral well Reservoir Pressure (psia) 1330 1220
spacing up to a critical value, however, recovery declines if Oil Gravity (°API) 7.1-11.5 10
the spacing exceeds this value. Closer well spacing provides a Oil Viscosity (cp) 6*104-1*107 9*105
higher rate of recovery in the early part of the project if steam Kinematic Oil Viscosity (ft2/day) - 0.0001076
Solution GOR (SCF/STB) 120-130 125
injected per injector is kept constant. When the total steam Wellbore Radius (inch) - 6
injection to the system is kept constant, there exists an Overburden Thermal - 34.6
optimum lateral spacing for the fastest oil recovery. Conductivity (Btu/ft/day/°F)
6. Low vertical permeability delays and reduces oil recovery Thermal Diffusivity (ft2/day) - 0.6566
by retarding the rise of steam chamber. Steam Injection Rate (STB/day) - 1000
7. Placement of the horizontal producer is more critical than Steam Temperature (°F) - 590
Steam Quality - 1.0
placement of the injector in a heterogeneous reservoir.
 A SIMULATION STUDY OF NOVEL THERMAL RECOVERY
SPE 76729 METHODS IN THE UGNU TAR SAND RESERVOIR, NORTH SLOPE, ALASKA 5

WELL % Oil
Recovery
80
75
20 70
10
20
60

75 50

40
30
20 30
---- 10
1000 1000
20

10
250
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Figure 1. Reservoir Geometry and Well Configuration Base Case
Time (thousand days)
EPSAGD EPSAGD-VI
SAGD-VI SAGD
Conv. Steam Flood

Figure 3. Comparison of Oil Recovery with Different Thermal

Processes in Ugnu Base Case Simulation

------- -------------
% Oil
Recovery
80
One vertical injector and a horizontal producer
70

60

-------------------- 50

40

Two vertical injectors at grid blocks 2 & 5 30

20

10
--- ----- ----

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Four vertical injectors at grid blocks 2,3,4,5 Time (Thousand Days)

Vertical Injector ------- Horizontal Producer Spacing 70 ft Base Case (10 ft)
Spacing 30 ft Spacing 45 ft
Figure 2. Well Patterns Used for SAGD-VI Process
(Top View)
Figure 4. Effect of Vertical Spacing on Oil Recovery
(SAGD)
6 B. SHARMA, S. KHATANIAR, S. PATIL, V. KAMATH AND A. DANDEKAR SPE 76729

% Oil % Oil
Recovery Recovery
80 80

70 70

60 60

50
50

40
40
30
30
20
20
10
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
Time (Thousand Days) 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Spacing 70 ft Base Case (10 ft) Time (Thousand Days)
Spacing 45 ft Spacing 30 ft
400 ft 300 ft 200 ft 100 ft
Figure 5. Effect of Vertical Spacing on Oil Recovery
(EP-SAGD) Figure 7. Effect of Lateral Well Spacing on Oil Recovery
(SAGD Process)

% Oil % Oil
Recovery Recovery
80 80

70 70

60 60

50
50

40
40
30
30
20
20
10
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (Thousand Days)
Time (Thousand Days)
2000 Days Preheat 1095 Days Preheat
365 Days Preheat SAGD (Base Case) 100 ft 200 ft 300 ft 400 ft

Figure 8. Effect of Lateral Well Spacing on Oil Recovery

Figure 6. Effect of Electrical Heating on Oil Recovery
(SAGD Process)
 A SIMULATION STUDY OF NOVEL THERMAL RECOVERY
SPE 76729 METHODS IN THE UGNU TAR SAND RESERVOIR, NORTH SLOPE, ALASKA 7

Oil
Production % Oil
(STB/day) Recovery
350 100

90 Layer 1
300
80
Layer 2
250 70

60
200
50
150 40

30
100
20
50
10

0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 .68/.68 .1/.68 .68/1
Time (Thousand Days) K(Layer 1)/K(Layer 2)

Kv/Kh = 1 Kv/Kh = .5 Kv/Kh = .15

SAGD EP-SAGD SAGD-VI

Figure 9. Effect of Anisotropy on Oil Production Rate Figure 11. Effect of Heterogeneity on Oil Recovery
(SAGD Process) (SAGD, EP-SAGD, & SAGD-VI)

% Oil
% Oil Recovery
Recovery 80
70
70
60
60
50
50

40
40

30 30

20 20

10 10

0 0
0 0.2 0.4 0.6 0.8 1 1.2 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Kv/Kh Time (Thousand Days)

SAGD EP-SAGD SAGD-VI K1/K2 = 1 K1/K2 = .147 K1/K2 = 6.8

Figure 12. Effect of Heterogeneity on Oil Recovery

Figure 10. Effect of Anisotropy on Oil Recovery (SAGD Process)
(SAGD, EP-SAGD, & SAGD-VI)
8 B. SHARMA, S. KHATANIAR, S. PATIL, V. KAMATH AND A. DANDEKAR SPE 76729

% Oil
Recovery
80

70

60

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Time (Thousand Days)

0.147 6.8 1

Figure 13. Effect of Heterogeneity on Oil Recovery

(SAGD-VI)